TheRock/compiler/amd-llvm/llvm/utils/update_mc_test_checks.py

#!/usr/bin/env python3
"""
A test update script.  This script is a utility to update LLVM 'llvm-mc' based test cases with new FileCheck patterns.
"""

from __future__ import annotations, print_function

from sys import stderr
from traceback import print_exc
import argparse
import functools
import os  # Used to advertise this file's name ("autogenerated_note").
import subprocess
import re
import sys

from UpdateTestChecks import common

mc_LIKE_TOOLS = [
    "llvm-mc",
]
ERROR_RE = re.compile(r":\d+: (warning|error): .*")
ERROR_CHECK_RE = re.compile(r"# COM: .*")
OUTPUT_SKIPPED_RE = re.compile(r"(.text)")
COMMENT = {"asm": "//", "dasm": "#"}

SUBSTITUTIONS = [
    ("%extract-encodings", "sed -n 's/.*encoding://p'"),
]


class Error(Exception):
    def __init__(self, test_info, line_no, msg):
        super().__init__(f"{test_info.path}:{line_no}: {msg}")


def invoke_tool(exe, check_rc, cmd_args, testline, verbose=False):
    substs = SUBSTITUTIONS + [(t, exe) for t in mc_LIKE_TOOLS]
    args = [common.applySubstitutions(cmd, substs) for cmd in cmd_args.split("|")]

    testline = testline.replace('"', '\\"')
    cmd = 'echo "' + testline + '" | ' + exe + " " + " | ".join(args)
    if verbose:
        print("Command: ", cmd)

    out = subprocess.run(
        cmd,
        shell=True,
        check=check_rc,
        stdout=subprocess.PIPE,
        stderr=subprocess.DEVNULL,
    ).stdout

    # Fix line endings to unix CR style.
    return out.decode().replace("\r\n", "\n")


# create tests line-by-line, here we just filter out the check lines and comments
# and treat all others as tests
def isTestLine(input_line, mc_mode):
    line = input_line.strip()
    # Skip empty and comment lines
    if not line or line.startswith(COMMENT[mc_mode]):
        return False
    # skip any CHECK lines.
    elif common.CHECK_RE.match(input_line):
        return False
    return True


def isRunLine(l):
    return common.RUN_LINE_RE.match(l)


def hasErr(err):
    return err and ERROR_RE.search(err) is not None


def getErrString(err):
    if not err:
        return ""

    # take the first match
    for line in err.splitlines():
        s = ERROR_RE.search(line)
        if s:
            return s.group(0)
    return ""


def getOutputString(out):
    if not out:
        return ""
    lines = []

    for line in out.splitlines():
        if OUTPUT_SKIPPED_RE.search(line):
            continue
        if line.strip("\t ") == "":
            continue
        lines.append(line.lstrip("\t "))
    return "\n".join(lines)


def should_add_line_to_output(input_line, prefix_set, mc_mode):
    # special check line
    if mc_mode == "dasm" and ERROR_CHECK_RE.search(input_line):
        return False
    else:
        return common.should_add_line_to_output(
            input_line, prefix_set, comment_marker=COMMENT[mc_mode]
        )


def getStdCheckLines(prefix: str, output: str, mc_mode) -> list[str]:
    o = []
    for i, line in enumerate(output.splitlines()):
        maybe_next = "-NEXT" if i > 0 else ""
        # Add an extra end-of-line check for --show-inst MCInst lines to
        # ensure we matched the full instruction name and not just a prefix.
        if line.startswith("# <MCInst "):
            line += "{{$}}"
        o.append(f"{COMMENT[mc_mode]} {prefix}{maybe_next}: {line}")
    return o


def getErrCheckLines(prefix: str, output: str, mc_mode, line_offset=1) -> list[str]:
    return [f"{COMMENT[mc_mode]} {prefix}: :[[@LINE-{line_offset}]]{output}"]


def parse_token_defs(test_info):
    tokens = {}
    current_token = None
    for line_no, line in enumerate(test_info.input_lines, start=1):
        # Remove comments.
        line = line.split("#")[0].rstrip()

        # Skip everything up to the instructions definition.
        if not tokens and not current_token and line != "//  INSTS=":
            continue

        if not line.startswith("//"):
            break

        original_len = len(line)
        line = line[2:].lstrip(" ")
        indent = original_len - len(line)

        if not line:
            current_token = None
            continue

        # Define a new token.
        if not current_token:
            if indent != 4 or not line.endswith("="):
                raise Error(test_info, line_no, "token definition expected")

            current_token = line[:-1].strip()
            if current_token in tokens:
                raise Error(test_info, line_no, f"'{current_token}' redefined")

            tokens[current_token] = []
            continue

        # Add token value.
        if indent != 8:
            raise Error(test_info, line_no, "wrong indentation for token value")

        tokens[current_token].append(line)

    return tokens


def expand_insts(tokens):
    def subst(s):
        for token, values in tokens.items():
            if token in s:
                for value in values:
                    yield from subst(s.replace(token, value, 1))
                return

        yield s

    yield from subst("INSTS")


def update_test(ti: common.TestInfo):
    if ti.path.endswith(".s"):
        mc_mode = "asm"
    elif ti.path.endswith(".txt"):
        mc_mode = "dasm"
    else:
        common.warn("Expected .s and .txt, Skipping file : ", ti.path)
        return

    triple_in_ir = None
    for l in ti.input_lines:
        m = common.TRIPLE_IR_RE.match(l)
        if m:
            triple_in_ir = m.groups()[0]
            break

    run_list = []
    for l in ti.run_lines:
        if "|" not in l:
            common.warn("Skipping unparsable RUN line: " + l)
            continue

        commands = [cmd.strip() for cmd in l.split("|")]
        assert len(commands) >= 2
        mc_cmd = " | ".join(commands[:-1])
        filecheck_cmd = commands[-1]

        # special handling for negating exit status
        # if not is used in runline, disable rc check, since
        # the command might or might not
        # return non-zero code on a single line run
        check_rc = True
        mc_cmd_args = mc_cmd.strip().split()
        if mc_cmd_args[0] == "not":
            check_rc = False
            mc_tool = mc_cmd_args[1]
            mc_cmd = mc_cmd[len(mc_cmd_args[0]) :].strip()
        else:
            mc_tool = mc_cmd_args[0]

        triple_in_cmd = None
        m = common.TRIPLE_ARG_RE.search(mc_cmd)
        if m:
            triple_in_cmd = m.groups()[0]

        march_in_cmd = ti.args.default_march
        m = common.MARCH_ARG_RE.search(mc_cmd)
        if m:
            march_in_cmd = m.groups()[0]

        common.verify_filecheck_prefixes(filecheck_cmd)

        mc_like_tools = mc_LIKE_TOOLS[:]
        if ti.args.tool:
            mc_like_tools.append(ti.args.tool)
        if mc_tool not in mc_like_tools:
            common.warn("Skipping non-mc RUN line: " + l)
            continue

        if not filecheck_cmd.startswith("FileCheck "):
            common.warn("Skipping non-FileChecked RUN line: " + l)
            continue

        mc_cmd_args = mc_cmd[len(mc_tool) :].strip()
        mc_cmd_args = mc_cmd_args.replace("< %s", "").replace("%s", "").strip()
        check_prefixes = common.get_check_prefixes(filecheck_cmd)

        run_list.append(
            (
                check_prefixes,
                mc_tool,
                check_rc,
                mc_cmd_args,
                triple_in_cmd,
                march_in_cmd,
            )
        )

    # find all test line from input
    testlines = [l for l in ti.input_lines if isTestLine(l, mc_mode)]
    # remove duplicated lines to save running time
    testlines = list(dict.fromkeys(testlines))
    common.debug("Valid test line found: ", len(testlines))

    # Where instruction templates are specified, use them instead.
    use_asm_templates = False
    if mc_mode == "asm":
        tokens = parse_token_defs(ti)
        if "INSTS" in tokens:
            testlines = list(expand_insts(tokens))
            use_asm_templates = True

    raw_output = []
    raw_prefixes = []
    for (
        prefixes,
        mc_tool,
        check_rc,
        mc_args,
        triple_in_cmd,
        march_in_cmd,
    ) in run_list:
        common.debug("Extracted mc cmd:", mc_tool, mc_args)
        common.debug("Extracted FileCheck prefixes:", str(prefixes))
        common.debug("Extracted triple :", str(triple_in_cmd))
        common.debug("Extracted march:", str(march_in_cmd))

        triple = triple_in_cmd or triple_in_ir
        if not triple:
            triple = common.get_triple_from_march(march_in_cmd)

        raw_output.append([])
        for line in testlines:
            # get output for each testline
            out = invoke_tool(
                ti.args.llvm_mc_binary or mc_tool,
                check_rc,
                mc_args,
                line,
                verbose=ti.args.verbose,
            )
            raw_output[-1].append(out)

        common.debug("Collect raw tool lines:", str(len(raw_output[-1])))

        raw_prefixes.append(prefixes)

    generated_prefixes = {}
    sort_keys = {}
    used_prefixes = set()
    prefix_set = set([prefix for p in run_list for prefix in p[0]])
    common.debug("Rewriting FileCheck prefixes:", str(prefix_set))
    ginfo = common.make_asm_generalizer(version=1)

    for test_id, input_line in enumerate(testlines):
        # a {prefix : output, [runid] } dict
        # insert output to a prefix-key dict, and do a max sorting
        # to select the most-used prefix which share the same output string
        p_dict = {}
        for run_id in range(len(run_list)):
            out = raw_output[run_id][test_id]

            if hasErr(out):
                o = getErrString(out)
            else:
                o = getOutputString(out)

            for p in raw_prefixes[run_id]:
                if p not in p_dict:
                    p_dict[p] = o, [run_id]
                    continue

                if p_dict[p] == (None, []):
                    continue

                prev_o, run_ids = p_dict[p]
                if o == prev_o:
                    run_ids.append(run_id)
                    p_dict[p] = o, run_ids
                else:
                    # conflict, discard
                    p_dict[p] = None, []

        # prefix is selected and generated with most shared output lines
        # each run_id can only be used once
        used_run_ids = set()
        selected_prefixes = set()
        get_num_runs = lambda item: len(item[1][1])
        p_dict_sorted = sorted(p_dict.items(), key=get_num_runs, reverse=True)
        for prefix, (o, run_ids) in p_dict_sorted:
            if run_ids and used_run_ids.isdisjoint(run_ids):
                selected_prefixes.add(prefix)

            used_run_ids.update(run_ids)

        # Use smallest outputs across RUN lines as sorting keys for
        # disassembler tests. Sort by instruction codes if no RUN line
        # produced a disassembled instruction.
        if mc_mode == "dasm":
            instr_outs = [
                o
                for prefix, (o, run_ids) in p_dict_sorted
                if o is not None and "encoding:" in o
            ]
            sort_keys[input_line] = min(instr_outs) if instr_outs else input_line

        # Generate check lines in alphabetical order.
        check_lines = []
        vars_seen = {prefix: dict() for prefix in selected_prefixes}
        global_vars_seen = {prefix: dict() for prefix in selected_prefixes}
        for prefix in sorted(selected_prefixes):
            o, run_ids = p_dict[prefix]
            used_prefixes.add(prefix)

            if hasErr(o):
                line_offset = len(check_lines) + 1
                checks = getErrCheckLines(prefix, o, mc_mode, line_offset)
            else:
                checks = getStdCheckLines(prefix, o, mc_mode)
                checks = common.generalize_check_lines(
                    checks, ginfo, vars_seen[prefix], global_vars_seen[prefix]
                )

            check_lines.extend(checks)

        generated_prefixes[input_line] = "\n".join(check_lines)

    # write output
    output_lines = []
    if use_asm_templates:
        # Keep all leading comments and empty lines.
        for input_info in ti.iterlines(output_lines):
            input_line = input_info.line
            if not input_line or input_line.startswith(COMMENT[mc_mode]):
                output_lines.append(input_line)
                continue
            break

        # Remove tail empty lines.
        while not output_lines[-1]:
            del output_lines[-1]

        # Emit test and check lines.
        for input_line in testlines:
            output_lines.extend(["", input_line, generated_prefixes[input_line]])
    else:
        for input_info in ti.iterlines(output_lines):
            input_line = input_info.line
            if input_line in testlines:
                output_lines.append(input_line)
                output_lines.append(generated_prefixes[input_line])

            elif should_add_line_to_output(input_line, prefix_set, mc_mode):
                output_lines.append(input_line)

    if ti.args.unique or ti.args.sort:
        # split with double newlines
        test_units = "\n".join(output_lines).split("\n\n")

        # select the key line for each test unit
        test_dic = {}
        for unit in test_units:
            lines = unit.split("\n")
            for l in lines:
                # if contains multiple lines, use
                # the first testline or runline as key
                if isTestLine(l, mc_mode):
                    test_dic[unit] = l
                    break
                if isRunLine(l):
                    test_dic[unit] = l
                    break

        # unique
        if ti.args.unique:
            new_test_units = []
            written_lines = set()
            for unit in test_units:
                # if not testline/runline, we just add it
                if unit not in test_dic:
                    new_test_units.append(unit)
                else:
                    if test_dic[unit] in written_lines:
                        common.debug("Duplicated test skipped: ", unit)
                        continue

                    written_lines.add(test_dic[unit])
                    new_test_units.append(unit)
            test_units = new_test_units

        # sort
        if ti.args.sort:

            def getkey(l):
                # find key of test unit, otherwise use first line
                if l in test_dic:
                    line = test_dic[l]
                else:
                    line = l.split("\n")[0]

                # runline placed on the top
                return (not isRunLine(line), sort_keys.get(line, line))

            test_units = sorted(test_units, key=getkey)

        # join back to be output string
        output_lines = "\n\n".join(test_units).split("\n")

    # output
    if ti.args.gen_unused_prefix_body:
        output_lines.extend(ti.get_checks_for_unused_prefixes(run_list, used_prefixes))

    common.debug("Writing %d lines to %s..." % (len(output_lines), ti.path))
    with open(ti.path, "wb") as f:
        f.writelines(["{}\n".format(l).encode("utf-8") for l in output_lines])


def main():
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument(
        "--llvm-mc-binary",
        default=None,
        help='The "mc" binary to use to generate the test case',
    )
    parser.add_argument(
        "--tool",
        default=None,
        help="Treat the given tool name as an mc-like tool for which check lines should be generated",
    )
    parser.add_argument(
        "--default-march",
        default=None,
        help="Set a default -march for when neither triple nor arch are found in a RUN line",
    )
    parser.add_argument(
        "--unique",
        action="store_true",
        default=False,
        help="remove duplicated test line if found",
    )
    parser.add_argument(
        "--sort",
        action="store_true",
        default=False,
        help="sort testline in alphabetic order (keep run-lines on top), this option could be dangerous as it"
        "could change the order of lines that are not expected",
    )
    parser.add_argument("tests", nargs="+")
    initial_args = common.parse_commandline_args(parser)

    script_name = os.path.basename(__file__)

    returncode = 0
    for ti in common.itertests(
        initial_args.tests, parser, script_name="utils/" + script_name
    ):
        try:
            update_test(ti)
        except Exception:
            stderr.write(f"Error: Failed to update test {ti.path}\n")
            print_exc()
            returncode = 1
    return returncode


if __name__ == "__main__":
    sys.exit(main())